Researchers from Radboud University and UMC Utrecht are developing a technology that will enable patients with speech disorders to communicate by converting brainwave data into words. The research predicts brainwave signals with 92 percent accuracy and translates them into words.
Researchers from Radboud University and UMC Utrecht have succeeded in converting brain signals into audible speech.
They used brain implants and artificial intelligence to map brain activity directly to speech in epilepsy patients.
This technology aims to give paralyzed and non-speaking people their voice back.
The researchers believe that the success of this project marks a significant advance in the field of Brain-Computer Interfaces.
Translated with www.DeepL.com/Translator (free version)
92 PERCENT ACCURACY RATE
The research, published in the Journal of Neural Engineering, used a combination of brain implants and artificial intelligence to capture spoken words with 92 percent accuracy.
While the technology currently focuses on individual words, future goals include the ability to predict full sentences and paragraphs based on brain activity.
According to lead author Julia Berezutskaya, a researcher at Radboud University’s Donders Institute for Brain, Cognition and Behavior and UMC Utrecht, the research marks a promising development in the field of Brain-Computer Interfaces.
“BY ANALYZING BRAIN ACTIVITY, WE CAN RESTORE PATIENTS TO THEIR VOICES”
“Ultimately, we hope to make this technology available for paralyzed patients who are unable to communicate. These people lose the ability to move their muscles and therefore to speak. By developing a brain-computer interface, we can analyze their brain activity and give them a voice again.”
Listening tests were also conducted with volunteers to assess how identifiable the synthesized words were.
The positive results from these tests show that the technology is successful not only in accurately identifying words, but also in communicating them in an audible and understandable way, just like a real voice.
“WE SEEM TO BE MOVING IN THE RIGHT DIRECTION”
“Our goal is to predict the exact sentences and paragraphs of what people are trying to say based only on their brain activity. To achieve this goal, we will need more experiments, more advanced implants, larger data sets and advanced AI models. All these processes will still take several years, but it seems that we are moving in the right direction.”